Last updated at Mon, 21 Aug 2017 19:11:52 GMT
I wish I had a dime for every time I have heard someone say “With so many vulnerabilities being reported in the Internet of Things, I just don't trust that technology, so I avoid using any of it." I am left scratching my head because these same people seem to have no issues running a Windows operating system.
As a researcher focused on the Internet of Things (IoT), I regularly release new vulnerabilities around IoT product ecosystems, which can include hardware, mobile application and cloud web servers, and APIs. The main goal when doing this work is better security. The hope is that the knowledge gained and shared in the course of this research will help manufacturers build better products and help those using IoT make better decisions around the deployment and management of IoT technology. Unfortunately, any vulnerability my peers and I discover during research can be—and often is— made out to be worse than it really is. Now don't get me wrong: I want the work I do to be taken seriously, but we still need to take a complete risk model into consideration when evaluating IoT issues. We can refer back to one of the most succinct risk formulas below:
Risk = Probability x Impact
Unfortunately, I think we have a tendency to forget about this formula. Every time I discuss my IoT findings with reporters, customers, or the general public, I like to discuss the associated risk, because it varies quite a bit based on “how” and “by whom” the affected IoT technology is being used.
A good example of this is the IoT tracking device research I conducted in the past year. If I told you a malicious actor could identify and use a simple low-energy Bluetooth dongle hanging on my keychain “used to find lost keys” to track me anywhere I go via Global Positioning System (GPS) data, what would you say about that? Our initial response would most likely be that this is horrible and scary, and that response is 100% understandable...but if we think through this using the risk formula, it can easily become both clearer and less dire.
First, what is the probability of someone's taking the time and effort to identify such a flaw in a product so they could track us? This leads to several new questions:
- Would anyone need or want to track me?
- Am I a very high-profile person, such as a politician or celebrity?
- Do I have any reason to believe someone is currently interested in stalking or tracking me?
If the answer is “no,” then the probability part of the equation quickly drops to nearly zero, or “very unlikely.” Remember basic math: If zero is multiplied by anything, it still equals zero—meaning the risk of this occurring is very low. Now if you do fall within the narrow category of users who may be at higher risk, then you might want to consider the benefit versus risk of using such technology, but for the most part few of us fall into that category.
If we talk through the risk of using IoT and apply the basic risk formula, we can better identify the true risk and, armed with that knowledge, focus on properly mitigating and/or reducing it.
When thinking about purchasing and deploying IoT at home or in business, there are more questions we can ask that can help us understand other aspects of the risks associated with IoT:
- Does the product support over-the-air security patching? Having the ability to quickly patch a product when vulnerabilities are found helps to reduce risk and is a key indicator of a more mature security program on the vendor's part.
- Has the vendor had an independent security review done on their product? Vendors who proactively test their products prior to going to market often deliver a more secure product; this is also another key indicator of a mature security process.
And, when we deploy our newly purchased IoT technology, what can we do as users/consumers to reduce risk?
- We should change default passwords and set complex passwords on all accounts and services (and choose devices/services that support this or even stronger authentication mechanisms).
- We should strongly consider deploying the solution within an isolated VLAN environment to protect our core business or home networks from any malicious traffic aimed at or coming from an IoT component.
I strongly encourage both businesses and consumers to embrace IoT and all the benefits that come with IoT solutions, but to do so with eyes wide open using common-sense risk evaluation, along with deployment and management best practices. This way, you can securely take advantage of all IoT has to offer.
Rapid7 can help you test the security of your IoT devices. Learn more about our IoT security services.